Article body


The famous debate between H.L.A. Hart and Lord Devlin was about principled and unprincipled criminalization. Hart argued that there was no principled justification for criminalizing many of the activities that Lord Devlin advocated criminalizing, such as homosexuality or prostitution. Hart argued for principled criminalization, which he suggested would be criminalization that could be justified by pointing to critical moral standards.[1] Hart took the view that culpable harm provided a critical moral justification for criminalization—that is, a justification that is universally right.[2] Joel Feinberg refers to critical morality as true morality, which according to him is “a collection of governing principles thought to be ‘part of the nature of things,’ critical, rational, and correct.”[3]

Like Hart, Feinberg asserts that a positive justification for criminalization is only valid to the extent that “it is also a correct rule of morality, capable of satisfying a transcultural critical standard.”[4] Feinberg attempts to limit criminalization by arguing that normative or objective moral accounts of harm can be used to constrain positive or conventional accounts of harm employed to justify penal censure. Critical moralists seem to take the view that deep personal conviction or practical reasoning allows moral agents to identify objective or normative accounts of harm.[5] I present a more modest account of the moral agent because I view the moral agent as nothing greater than a communally situated human being. If practical reasoners are merely communally situated humans trying to solve conventional conflicts, then it is fairly clear that it is impossible for such creatures to identify fully “correct” accounts of harm and badness, goodness, rightness, wrongness, and so forth.

Standards identified by human thinkers cannot be truly correct because it is impossible for us to know whether a standard is truly correct. Furthermore, in practice all reasoning (notwithstanding the belief of some commentators that these standards are mind-independent) is influenced by societal evolution, convention, and human biases.[6] The human mind is not a computer! It is not possible to claim that certain conventional wrongs are truly wrong, bad, or harmful in a transcultural sense, but principled justifications can be supplied for criminalizing many conventionally contingent wrongs. There is no doubt that intersubjective deliberation will give us better results, but it cannot tell us whether a particular moral standard is correct.

Hart suggests that an objective account of harm can be discovered and therefore can provide a critical moral justification for or against criminalization. But the harm principle itself is a conventional construct, and conceptualizations of harm depend on convention. It might be argued that deep (“deep”, meaning long-held and widely shared understandings in Western society) conventional agreement about the harmfulness of certain acts, such as murder, is sufficient to provide a principled-harm argument for outlawing it. This provides a strong conventionally objective case for outlawing such wrongs. Feinberg, however, supplements the harm principle[7] with an offence principle,[8] which holds that culpable offence-doing also provides a critical moral justification for criminalization. The problem with offensive conduct and trivial harms is that there is no deep or constant (intersubjectively shared) agreement about the badness or wrongness of such acts. I could provide many examples, but I think nudity in ancient art, movies, and modern art, might be sufficient for tentatively claiming that exhibitionism has not been constantly considered to be bad, harmful, or wrong. Feinberg is particularly critical of Lord Devlin’s positive morality,[9] but it is not clear that Feinberg’s offence principle rests on anything more than positive morality. Feinberg does not explain why culpable offence-doing is inherently wrong in a critical moral sense rather than a conventional sense—or why standards cannot be developed from conventional morality to provide principled justifications for criminalization.

While I develop the idea of conventional objectivity more fully throughout the paper, the basic theory is that we are able to draw on our deeply held conventional understandings of wrong and harm (including our scientific and biological accounts of harm and bad consequences—in addition to conventional understandings about privacy and autonomy in modern society) in order to formulate a case either for or against criminalization. We may change our minds about what is bad, harmful, and wrong depending on the social context. Hence, our conceptualizations of harm and wrong depend on conventional understandings of harm and on socialization. We may therefore claim that something is objectively harmful within a certain conventional context, but this is entirely different than claiming that something is bad or harmful in a transcultural critical objective sense. For the most part, at the most basic level all societies have similar conventional understandings about the badness, wrongness, and harmfulness of conduct such as genocide, murder, starvation, torture, and so forth. Such understandings have emerged because humans have drawn on basic biological information, human instincts, and evolving social norms to solve conventional conflicts.

Transculturally, there are shared understandings about the badness and harmfulness of fairly primitive[10] harms such as wantonly amputating another’s hand. For instance, in some countries the justification for chopping off a thief’s hand for shoplifting hinges on an understanding that it is bad and harmful to wantonly amputate a person’s hand. It is because hand amputation is understood to be bad and harmful that it is used as a punishment rather than a reward. I do not know of any state where the conventional understanding is that hand amputation is good and thus should be used as a reward. The same might be said for the death penalty. There is no transcultural disagreement about death (capital punishment) or hand amputation as bad and harmful. Rather, the disagreement is about whether such punishments are proportionate or necessary given our respect for humanity and life. But this does not mean those acts are truly harmful or bad. Empirical information (i.e., biological, scientific, and medical explanations of pain and damage) and our conventional understanding of pain, hurt, and culpability, are more than sufficient for providing an objective account of the harmfulness of wanton hand amputation; this alone, however, cannot be used to prove that it is objectively harmful in a critical moral sense.

Ashworth’s claim that the criminal law has been influenced by the political demands of the day is beyond dispute, and unless we can identify appropriate constraints, it might be impossible to have a principled criminal law.[11] If the harm and offence principles do not provide critical reasons for constraining criminalization, then it might not be possible to distinguish Feinberg’s justifications for criminalization (culpable harm and culpable offence) from those of Lord Devlin. I argue that both Hart and Feinberg were wrong to assume that there is a critical moral type of harm and a conventional moral type of harm. I take the view that culpable harm and culpable offence only provide conventional justifications for criminalization. I focus on Feinberg’s offence principle because the conventional nature of culpable offence-doing provides the strongest challenge to the claim that criminalization can be constrained by critical moral accounts of harm and offence. It is difficult to see the critical objectivity of wrongness claims concerning many forms of offence-doing, such as public exhibitionism. The aim of this paper is to show that offence to others does not provide a critical moral justification for invoking the criminal law.

J.L. Mackie used “retribution” as a test case for objectivity.[12] He would have had a field day with Feinberg’s offence principle. I am a supporter of Feinberg’s criteria, but I think the metaethical foundations that he claims for his harm and offence criteria are open to question. Feinberg promises a normative conception of wrong distinct from the positive one upon which Lord Devlin relies, but “offence to others” delivers a conception that is indistinguishable from a merely positive one. If harm or offence is anything that a person subjectively perceives to be harmful or offensive, then Feinberg’s principles are vacuous. To counteract this possibility, Feinberg argues that the harm or offence must be objective or normative. Feinberg seeks to base his harm and offence principles on objective foundations, but fails. When it comes to the offence principle, the weakness of Feinberg’s critical objectivity claim is most evident. I argue that even though there is no critical moral justification for criminalizing exhibitionism, it is possible to draw on conventional morality to provide a principled case for criminalizing it in certain contexts.

I. Criminalization

The criminal law gives any government immense power over its people because disobeying a criminally codified command can result in stigmatization and severe punishment (conviction, imprisonment, fines, and so forth). Why are citizens of a given state, bound or obliged to follow the guides of acceptable behaviour as set out in the criminal law? Since the criminal law is a punitive response to unwanted behaviour, its authority is contingent on its legitimacy.[13] I argue that certain human acts are deserving of the crime label because they produce bad consequences (or risk producing bad consequences, as is the case with attempts, endangerment, and so forth) that are of an avoidable kind for others—avoidable in that the wrongdoer culpably aimed for the bad consequences and could have chosen otherwise. The harm might be indirect and thus threaten the community (State) by damaging its institutions (e.g., perjury, bribery, and environmental damage are collective harms), or the harm might be directly victimizing, as is the case with murder, theft, rape, and so forth. The issue of indirect or collective harm is controversial as it is difficult to individualize the harm-doing.[14]

Many criminal laws are codified and deeply held conventional commands, such as laws against rape, assault, murder, theft, fraud, and so forth. A given criminal law will have authority regardless of whether it serves a legitimate purpose,[15] but the criminal law as a general institution of social control will retain its legitimacy and authority only if the bulk of its commands are understood as principled—that is, understood to be fair in accordance with our conventional understandings of justice and fairness. It is not possible to state the cut-off point in numerical terms, but if more than fifty percent of a given state’s laws served no legitimate purpose (that is, some goal that is understood to be legitimate by communally situated moral agents), or were unjust and draconian, then the result might be revolution. For instance, most people in advanced Western societies would not tolerate jail terms of fifty years for shoplifting.[16] We have deep conventional understandings about the trivial nature of the harmfulness of shoplifting and therefore do not see lengthy jail terms as a necessary government response. When the bulk of a state’s laws, whether they are private laws or public laws, serve some legitimate purpose (purposes that are conventionally understood and accepted as legitimate), the law in that state will retain its posited authority. Although there are many perceivably unjust criminal laws that serve no legitimate purpose,[17] the bulk of jailable offences in the United States do seem to be aimed at genuine wrongs.[18] Arguably, many unjust laws retain their authority because the general institution of criminal law and punishment retains its authority.

We implicitly agree to have law in order to maintain society for the good of humanity, but we also realize that the State might misuse the law or simply get it wrong. Thus, the law itself has to be subject to a number of constraints in the interest of fairness. The State is merely comprised of members of society acting as a collective—and we need to know why the commands of the majority, as expressed in laws, have authority over us as individuals. As we will see, many activities that are deemed bad or harmful by the State (the collective, community, or society) often involve little more than someone flouting a seemingly innocuous social custom, as is the case with exhibitionism. Since this type of conduct is harmless, the harm criterion does little to answer the following question: what makes exhibitionism criminalizable?

Criminalization is a process of labelling certain actions as punishable by the State in order to solve social conflicts and problems with co-operation that arise in competitive plural societies.[19] I argue below that constraints, such as those of harm and culpability, are only objective to the extent that there is deep conventional agreement about what constitutes a punishable harm. However, once we get into territory where there is disagreement about what ends are intersubjectively shareable by all communally situated agents, a principled case for criminalization is difficult to identify. Furthermore, it is almost impossible to identify a critical moral account of criminalizable harm and offence. At the inter-jurisdictional level there is deep agreement about the badness and wrongness of acts that result in primitive harm, such as gross physical harm (e.g., biologically painful harms such as starvation, blinding, amputation, and torture). Beyond those primitive harms, however, agreement is totally contingent on jurisdictional and cultural conventions. Conventional harms and conventions are born and must die together.[20] Primitive harms do not rely on convention as there are other biological and scientific explanations of the badness and harmfulness of such acts. We could adopt a grander scheme and argue that the normativity of harm hinges on critical agent-relative reasons. [21] As Christine Korsgaard puts it, “Values may be intersubjective: not part of the fabric of the universe or external truth, but nevertheless shared or at least shareable by agents.”[22] Intersubjectively, “normative claims are not the claims of a metaphysical world of values upon us: they are claims we make on ourselves and each other.”[23]

But Korsgaard has a Kantian inter-subjectivity of all reasonable agents in mind. The word “shareable” reflects Kant’s unwillingness to engage in anthropological reasoning when thinking about the content of morality. It is impossible to claim that criminal law is normative in this sense because anthropological information has to be considered in light of the conventional problems it addresses.[24] Principled criminalization might be identified by examining what is actually shared in specific communities, but this would not be normative in the Korsgaardian sense as it is contingent on what is shared in specific contexts and at specific points in time.

Culturally situated intersubjective agents might identify principled justifications for criminalization, but there would be nothing critically objective about the standards that they might develop. Furthermore, my conception of “principled” is neutral between different reasons why agents might intersubjectively share their ends—and is therefore neutral between what I describe as the critical moralist’s reason for arguing that ends should be shared and other reasons for so arguing. I merely critique critical morality to highlight the impossibility of claiming that particular acts, such as exhibitionism, are objectively wrong in a critical moral sense and are therefore prima facie criminalizable. The exhibitionism exemplar is used in the final sections to emphasize the vacuity of the claim that certain disgust-causing acts can be defined as universally wrong and bad.[25]

It is fairly easy to show that core instances of criminality (rape, theft, murder, and so forth) are principled if we accept deeply held conventional conceptualizations of harm, autonomy, and culpability. There is much less agreement, however, when it comes to criminalizing offensive conduct and soft harms.[26] A principled conventional account of the badness of offending others might be possible, but it is impossible to provide a critical moral account of its inherent criminalizability. The best we might be able to do is accept that Lord Devlin was right to argue that everyone (communally situated) intersubjectively shares the end of social harmony, but that everyone might also share the end of social toleration when social harmony is merely disturbed by tolerating conventionally harmless conduct that involves the fundamental liberty and equality interests of those causing the offence. For instance, homosexuality and prostitution are consensual activities that take place between adults behind closed doors, and criminalizing either would violate the privacy, equality, and autonomy rights of the offenders.

I argue that unexplained claims of objectivity or critical morality are not sufficient to refute Lord Devlin’s theory. Per contra, an intersubjectively constrained conventional morality might explain the difference between good reasons for criminalization and those that purely cater to idiosyncratic prejudice.[27] The critical moralist does not merely aim to subject justifications for criminalization to critical scrutiny, but also claims that their reasons provide correct, transnational, or universal justifications for criminalization. These claims overlook the fact that the wrongness and badness of the acts (and consequences that flow from certain social interactions) are circumstantially and conventionally contingent. I cannot see why a strong conventional account of culpable harm and offence-doing is not sufficient to provide a principled account of their criminalizability. It does, however, acknowledge that what presently seems to be a good justification for criminalization might not be so later. The core problem for lawmakers is that once they move away from accounts of harm and wrong where there is almost omnipresent social agreement about the harmfulness of the act, as is the case with gross physical harms (i.e., primitive harms such as, physical starvation, blinding, wounding, and so forth), the harmfulness of the conduct becomes conventionally contingent—and if we move far enough away from the harm paradigm, then it becomes impossible to even describe the unwanted conduct as conventionally harmful.

If we can imagine a bad-consequence dartboard of unwanted conduct, primitive harm would be the bull’s eye. As we move concentrically away from the bull’s eye, the harms become more conventionally contingent, and ultimately the unwanted consequence is not harmful at all, but a mere flouting of some custom or social norm. The basic elements of wrongness for the purpose of criminalization are: bad acts and consequences, and culpability. It is conventionally understood that the union of these makes an agent’s actions wrong and ultimately criminalizable. Culpability is about telling an agent in advance, “If you aim or disregard an obvious risk for a particular bad consequence, then you will be punished for your choice.” The bad act and consequence constraint is not too controversial when it is actual harm, or risk of harm in the case of attempts and endangerment. However, people disagree about whether offences and soft harms also constitute harm. Conventionally, there is deep agreement about the legitimacy of criminalizing wrongful harm. There might also be agreement about the need to regulate conventionally contingent wrongs, such as exhibitionism, but what is clear is that a critical moral or critically objective account of the wrongness of offending others is vacuous.

If Hart, Feinberg, and countless others want to dismiss Lord Devlin’s positive morality, then they must show why their accounts are different. Furthermore, their claim that only critical moral conceptualizations of harm and offence provide principled justifications for criminalization, is nonsensical because they have not shown why their accounts of harm and offence are critical. I argue that conventional harms and offence, as identified intersubjectively by communally situated deliberators, is sufficient to scrutinize criminalization decisions and to identify a principled case for criminalization.

II. The Vacuity of Critical Moral Accounts of Harm and Offence

What are the moral aims of the criminal law? The object and function of law generally is not too different from that of conventional morality. Mackie provides a superlative précis of the function of morality and its relation to law:

Protagoras, Hobbes, Hume, and Warnock are all at least broadly in agreement about the problem that morality [and ultimately law] is needed to solve: limited resources and limited sympathies together generate both competition leading to conflict and an absence of what would be mutually beneficial cooperation.[28]

Mackie also explains that

[t]he essential device [for creating society and co-operation] is a form of agreement which provides for its own enforcement. Each of the parties has a motive for supporting the authority who will himself have the job of punishing [or awarding private law remedies such as damages, injunctions, and so forth] breaches of the agreement (and will himself have a motive for doing so). Consequently each party will have a double reason for fulfilling his side of the bargain: the fear of punishment [or having to pay damages, for example] for breaking it, and the expectation of benefits from keeping it, because the fulfilment by the [majority of] other parties of their sides of the bargain is fairly well assured by the same motives.[29]

Whether we are talking about morality by agreement[30] or the social contract more generally, [31] there is ample empirical evidence to support the claim that society is formed by some kind of agreement,[32] and also that some individuals will not keep their side of the bargain in such a big web of complex agreements and inter-agreements.[33] Consequently, informal moral commands are codified into law so that violations will be deterred with punishment or private law remedies.[34] Mackie cites game theory in his discussion of the evolution of morality, but he is careful to note that even the most advanced theory could not explain the complexity of the way in which moral principles have evolved from the process of human socialization and civilization.[35] We benefit from aviation, telecommunications, university education, and travel; that is, from property, and services, and the laws that are designed to regulate the fair distribution of such goods and services. It is in these areas that the law is also needed to prevent harm to others. For example, health and safety standards, and regulations against fraud and deceptive practices, are designed to reduce harm. Since the State and its institutions advance co-operative living and, ultimately, human flourishing, each individual has an interest in maintaining them. As we will see, modern accounts of morality have deprived the intersubjective thinker of a social milieu.[36]

Raz notes that law serves a number of social functions including the prevention of undesirable behaviour mainly achieved by enacting criminal and tort laws through the provision of facilities and mechanisms to allow private arrangements to be regulated and protected between individuals; through the provision of services and the redistribution of goods; and through the provision of facilities for solving unregulated disputes.[37] Since society is necessary for the advancement and well-being of humanity, it is maintained both directly and indirectly by law. Laws cover many areas because of the complexity of modern living. We have criminal law, contract law, family law, trust law, consumer protection law, tort law, environmental law, tax law, and so forth. Tax law, for example, has both a direct and indirect impact. It forces individuals to hand over a portion of their income, but that income is spent on communal infrastructure. Tax law allows revenues to be collected in a transparent way so that the public may benefit indirectly from the provision of universities, schools, roads, courts, police, welfare for the poor, and so forth. The provision of these services reduces conflicts that might arise from the extreme distribution disparities that flow from inability.[38]

Principled criminal laws should be formulated by drawing on rationally constructed principles of justice, that is, principles that have evolved from deeply held conventional understandings of justice and fairness. Principles of justice such as the harm principle, the autonomy principle,[39] the culpability principle,[40] and the equality principle,[41] among others, have been developed and constructed by humans, and have improved as humans have gained better insights. Of course, accounts of harm will vary given the limits of epistemological inquiry and of human rationality. Human agents invent crimes to manage conventional conflicts that arise from communal living. Criminal law is a system of social control that allows a given community to manage itself.[42] It is used to manage genuine conflicts, but unfortunately, also to criminalize conventionally harmless wrongs and to control less powerful groups in society.[43]

Criminal laws that proscribe wrongs that are not harmful, or do not violate the autonomy of others, are unprincipled. For example, the acts referred to by Lord Devlin[44] are not conventionally harmful or oppressive to the autonomy of others because even conventional accounts of harm and wrong cannot explain how consenting adults engaging in homosexuality or prostitution could harm or violate the autonomy of others. Lord Devlin does not run into error by suggesting that without criminalization of these acts, society would disintegrate, but rather he runs into error by postulating that certain harmless violations of conventional norms would cause social disintegration and thus should be criminalized. There is no empirical support for his claim that activities such as homosexuality or prostitution would cause the same type of social disintegration that would transpire if wrongful harms such as murder, rape, theft, and robbery, were not criminalized.[45]

Rational deliberators should draw on the best social information available—including deep conventional understandings of justice, harm, privacy, autonomy, and so forth—when making criminalization determinations. The evolution of criminal law has often been shaped by unjust considerations because of lawmakers who were not sufficiently enlightened and rational at various stages in our history to understand the injustice of some of their decisions.[46] In the sixteenth and seventeenth centuries, the masses lacked the context to understand that humans could not really be witches, and therefore many women were criminalized for allegedly engaging in witchcraft.[47] We no longer criminalize witchcraft as we have sufficient empirical information to be able to rationally understand that humans cannot have supernatural powers. The issue of objectivity, even in the limited conventional sense, is fundamental as it can explain the wrongness of actions such as genocide, murder, rape, and so forth.[48] Reason allows intersubjective thinkers to see that the gross physical harm-doing involved in culpable genocide is objectively bad and wrong, regardless of the context or circumstances. Conventionally, it is understood as a gross and wanton abuse of human life. The same deliberator would also understand that the wrongness of exhibitionism is conventionally contingent—to ascertain its badness and ultimately its wrongness, the deliberator also has to consider the underlying social norms that inform it.

Wrongness grounded in critical morality is inherent wrongness—that is, those wrongs that are truly wrong. Objectivity here, claims that the proposition “X is wrong” is an absolute truth. Wrongness that is supposedly discovered as a truth (ethical wrongness grounded in moral and epistemological realism) is distinguishable from wrongness that is derived from communally situated agents intersubjectively reflecting on evolving standards of justice. The latter considers how your culpable actions will impact the interests of others in certain social contexts. The union of bad acts and consequences with culpability, as it is conventionally understood, is sufficient for establishing wrongness and thus a conventionally objective case for criminalization. This case for criminalization may not be objective in a critical moral sense, but it may be the best that we can do. A communally situated moral agent can act rationally and can be a detached observer who is appraised of the principles of justice that have evolved (such as the harm principle, the culpability constraint, and so forth), and the relevant social facts and conventions; and thus this agent can be in a position to reason and understand that certain culpable actions are wrong and worthy of punishment.

The conventional account is more constructive in criminalization ethics because it allows the theorist, philosopher, politician, and citizen to draw not only on abstract concepts such as justice, autonomy, harm, fairness, equality, and humanity that have been thought about and developed by thinkers for generations,[49] but also empirical information, context, convention, social practice, and so forth, in order to formulate practically useful guiding principles for constraining unjust criminalization in competitive societies. The reflective endorsement approach is about applying the criminalization label to violations that humans can reason are wrong because of their impact on genuine human interests in organized, co-operative, coordinated, and civilized societies. The constraints against unprincipled criminal law might include criteria such as harm and culpability. Critical moral accounts of harm differ in that such harms are always harms regardless of the time or context.

The most extreme claims of objectivity or normativity come from the moral realists who claim that certain actions are wrong in a mind-independent way—that is, wrong regardless of whether there are humans (including socially conditioned humans) available to conceptualize their wrongness.[50] I propose that it is nonsensical to argue that the consequences of death, physical pain, or harm are bad consequences in thought-independent terms, not only for humans but also for animals, trees, and all life forms on the planet. The ontological idea that the consequence of death or physical harm to a life form really would exist, and would in fact be bad in strong mind-independent terms, is oxymoronic because it relies on human preconceptions of the “what if”. The bad consequences that would allegedly exist independently of human thought, such as an earthquake wiping out a species, are only bad according to a human conceptualization of “bad”. The realist claim is that it is not that this would not be bad without humans, but that it would exist as something different. Maybe it would have a different label, but it would be the exact same physical set of events. Furthermore, wrongness is a human construct that rests on culpability (mens rea)—that is, human intentions. Animals instinctually avoid harm and death. Even if humans did not exist and could not conceptualize a snake biting and killing an elephant, merely because it erroneously feared that the elephant was going to stand on it, the death of the elephant would exist. However, the snake cannot be culpable. Thus, it can harm the elephant (harm as conceptualized by humans) but it cannot wrong it, because it cannot know any different. Putative self-defence might justify a human acting as the snake did, but a snake does not have the capacity to comprehend wrongness and thus does not need to defend its actions. Likewise, a volcano might harm a species by wiping out the rainforest on which it depends for food, but the volcano does not thereby wrong those creatures.

Domestic cats have a tendency not only to kill birds and mice for food, but also to torture such creatures by playing with them for many hours before eating them. In some cases, the cat will not even eat the bird or mouse, but will merely use it for the fun of playing with it. When a human sees a bird or mouse being tortured as a cat plays with it, the tendency is to try to rescue the prey—especially if it is a bird—due to conventional norms about birds being good and mice being vermin. The human intervener sees the cat’s wanton use of its prey as bad. However, no one would consider punishing the cat, as rational humans realize that a cat does not have the reflective and rational capacities of a human being and therefore does not bring about the bad consequences culpably.[51] Per contra, when a person intentionally aims to bring about avoidable bad consequences for others, it is the person's moral culpability and the badness of the consequences (harm to a fellow human being) that provides the lawmaker with a conventional justification for criminalization. It is fair to punish those who deliberately harm others because harm-doing produces bad consequences for those who are harmed, and the harm-doer knows that they are committing a wrong by inflicting such harm. It violates the genuine rights of the victims.

A more sophisticated realist argument is that certain acts are wrong in a mind-independent sense. Science-based ontology might be useful for claiming that biological harm such as blinding a human, amputating their legs, or subjecting them to a lobotomy, for example, is truly damaging and painful in an ontological and scientific sense, and thus bad. But how could intentional human actions (blameworthy actions, such as those involving culpability) be mind-independent? Surely the intentional harm-doing has to be carried out by a creature of human intelligence with a mind that is in operation in order for it to be willed and intended. It is our conventional conceptualization of culpability and harm that is doing all the work in these moral theories. When a human thinks, plans, deliberates, and then harms others, the willed harm could hardly be mind-independent. It can only be understood as wrong if there is a human knower to grasp its wrongness. It is wrong because a creature (a human being) that has enormous intelligence, and has evolved and socialized itself for millennia,[52] is able to draw on its intellect, rational capacity, social convention, empirical and biological facts, and conventional understandings,[53] in order to realize the wrongness of intentionally harming others.

When conflicts or clashes arise between human agents, the same intersubjective agents reflect to determine which party is intentionally, or recklessly, acting unjustly—that is, committing a wrong. For instance, the idea of queuing for customer service is a convention that evolved to solve the conflict that would arise if everyone tried to be served at the same time. Likewise, the culpability constraint evolved[54] from the reflective endorsement process as it became clear that intentionally or recklessly[55] aiming to bring about avoidable (culpable) bad consequences for others was different from accidentally doing so. The deliberator does not have to reflect too deeply to understand that those who fail to queue without excuse or justification act at the expense of all those who have.

To summarize the foregoing, a conventional approach[56] acknowledges that acts are only wrong for humans if humans conceptualize them as being wrong, bad, and harmful—and that for humans to do this, they must draw on social and contextual information. Blame and fault are conventional concepts that have evolved from human rationality—that is, human reasoning about fairness and justice, and the related institutions and social practices that have evolved as humanity has become civilized and socialized. It is unproductive to attempt to demonstrate that criminal wrongs are objectively wrong in the critical sense.[57] Moral principles such as the culpability principle are instantiated in the world, but are human constructs. The culpability condition requires a mind of some kind (mind-dependence rather than mind-independence) and rationally groundedsocial transactions. A cat torturing a mouse is not a rationally grounded social transaction because cats lack rationality and operate outside our social milieu. Soldiers, by contrast, are rational human agents who are able to draw on principles of justice and social norms in order to engage in rational social transactions. Consequently, if a group of soldiers were to torture prisoners of war, they would be acting irrationally and would also contravene deeply held social rules about not torturing others. The soldiers have sufficient rationality and empirical information to understand the relevant social information and conventional implications of their actions so as to identify the wrongness of torture.

Even if it is possible to determine the absolute truth of certain moral propositions about the inherent wrongness of certain crimes or the metaphysical status of offending others, there are many crimes, such as exhibitionism, that cannot be explained as having truly bad consequences for all people at all times.[58] I have not seen a convincing account of the truth of the proposition that being naked in public is truly wrong in an inherent universal sense. It is not a mere case of whether offence and disgust are properties that are instantiated in the world,[59] but whether exhibitionism does in fact produce an inherently bad consequence. Socialization seems to provide the better explanation of the disgust-causing properties of public nudity. There is nothing inherently wrong with the nudist using a public beach—one hundred years ago, wearing a modern bikini in public would have been the equivalent of being nude today, and one hundred years from now nudity might be the norm on beaches. Nevertheless, as I point out below, we might regulate public nudity for the sake of solving co-operation problems concerning the ethical use of public spaces in complex, plural societies. Critical moral accounts of the badness of offending others—such as that provided by Feinberg—have failed to demonstrate the inherent wrongness of public nudity, because outside of human thought, socialization, context, and convention, it does not produce a bad consequence and is not absolutely wrong in a universal sense.[60]

It is not only the wrongness of offensive acts that is conventionally contingent since genuine harms are also conventional. For instance, if X were to paint a yellow stripe across the Mona Lisa, X’s conduct would be classified as criminal.[61] But unless we consider the underlying social norms, it is not possible to comprehend the wrongness, badness, or harmfulness of intentionally painting an additional feature on an old painting. Let us assume that X is a private collector, so conventional property rights are not violated. If X owns the painting, then surely X is entitled to destroy it. Some might argue that the additional paintwork is further art and adds dimension to the original artwork. It certainly does not diminish the owner’s essential or primitive-type survival resources in the way that destroying a remote community’s only source of water and food would.[62] The objective wrongness of conventional harms and offences can only be ascertained by considering contextual, circumstantial, social, and empirical factors. Therefore, the objectivity of this type of harm is conventional, and thus is subject to all the inconsistencies and biases that affect the reasoning of communally situated intersubjective agents. Remember, my communally situated agents do not have the supernatural capacity envisaged by Nagel, Kant, or Korsgaard. They are just socialized humans drawing on societal practices to try and work out what conventional values should be protected through criminalization. In this sense, it is necessary to understand conduct in light of the social norms that inform it.[63]

Similarly, if a person takes a coin and scratches the paintwork on another person’s new Rolls-Royce, the car owner has been harmed in a conventional sense. However, if it is a minor scratch, it seems that the car owner has only been offended (rather than harmed), because the owner has been socialized to enjoy the aesthetics[64] of cars with perfect paintwork. The shallow scratch would not need to be repaired, as the car would not rust, nor would it affect the car’s usability. The reality in modern complex societies is that many genuine harms and offensive wrongs can only be understood by considering the underlying social norms, as bad acts or consequences have a substantial man-made element. These bad consequences occur as a result of the complex way in which we have socialized ourselves, and because we agree to have our freedom constrained in certain social contexts in order to achieve the levels of co-operation that are essential for society, community, and civilization to function and exist.

In what follows, I will outline the vacuity of Feinberg’s claim that culpable offence-doing provides a critical moral justification for criminalization. The harm principle is less problematic because there is deep conventional agreement not only about the most primitive harms, but also about many harms that are not primitive such as destroying cultural artifacts like the Mona Lisa.[65] The core issue in the harm principle context will be to ensure the harm is genuine and that the criminal law is a proportionate legislative response.[66] I think the offence principle is much more controversial as it is not clear that preventing offence, per se, is needed to promote human flourishing, or that it solves conflicts that need to be solved by the criminal law. I also examine whether Feinberg’s offence justification for criminalization can be distinguished from Lord Devlin’s. I set the scene by briefly discussing the conventionally contingent nature of harm.

III. Conventionally Contingent Harms

Wrongs emerged naturally as society became more complex and as more intricate problems with co-operation arose. Experience taught people that it was necessary and good to avoid harms and other bad consequences, especially those of the culpable kind. Let us consider conventional harms in light of Feinberg’s harm principle. Feinberg expounds harm in three senses: (1) harm as damage, (2) harm as a setback to interests, and (3) harm as wrongdoing.[67] Harm, as used in Feinberg’s formulation of the harm principle, is an amalgamation of senses (2) and (3). Harm must be caused by wrongful (culpable) conduct to be a candidate for criminalization. Harm occurs under the harm principle when X’s interests are set back by the wrongful conduct of Y.[68] The concept of harm as used by Feinberg represents “the overlap of senses two and three: only setbacks of interests that are wrongs, and wrongs that are setbacks to interests, are to count as harms in the appropriate sense.”[69] When used in this way, the term “interest” refers to a stake that a person has in his or her well-being. According to Feinberg, one’s interests taken as a whole consist of all those things in which one has a stake. In the singular, one’s personal interest “consists in the harmonious advancement of all one’s interests in the plural.”[70] These interests, or as Feinberg puts it, “the things these interests are in, are distinguishable components of a person’s well-being: he flourishes or languishes as they flourish or languish.”[71]

Feinberg explains the badness of harm-doing by referring to a trichotomy of interests, including welfare interests, and those security and accumulative interests that cushion our welfare interests.[72] Welfare interests are at the core of Feinberg’s scheme. They are interests of a kind shared by almost everyone as “necessary means to ... more ultimate goals, whatever the latter may be, or later come to be.”[73] Welfare interests include our interest in prolonging the continuance of our life for a foreseeable period of time, preserving our physical health and security, maintaining minimum intellectual acuity and emotional stability, being able to engage in social intercourse and benefiting from friendships, sustaining minimum financial security, sustaining reasonable living conditions, avoiding pain and grotesque disfigurement, preventing unjustified anxieties and resentments, and being free from unwarranted coercion.[74] They are those interests in goods and conditions that we all need, independently of our individual life-plans. Everyone has a necessary stake in these kinds of interests as they are the requisites of our well-being. [75]

Feinberg distinguishes important welfare interests from those interests that merely concern a person’s more ulterior aims.[76] Our ulterior aims might include the goal to own a dream house, or to have a prominent career as a movie star or as a politician, and so forth.[77] A person’s more ultimate goals and wants (e.g., building a dream house, gaining a political or professional position, solving some vital scientific question, raising a family, or achieving spiritual grace) are not directly protected by the law:

If I have an interest in making an important scientific discovery, creating valuable works of art, or other personal achievements, the law will protect that interest by guarding my welfare interests that are essential to it. But given that I have my life, health, economic adequacy, liberty, and security, there is nothing more that the law (or anyone else, for that matter) can do for me; the rest is entirely up to me.[78]

Ulterior interests that extend elements of welfare beyond minimal levels, however, are also protected.[79] The law against burglary not only protects the welfare of the indigent person who might face starvation if burgled, but it also protects the billionaire whose welfare might not be directly affected by the theft of a Caravaggio painting that they forgot they owned.[80] Even though certain types of harm only have a trivial impact on the interests of certain individuals, they can have an accumulative impact. Hence, it is not only the ulterior interests of billionaires that are protected,

but also their interests in liberty (the interest in being the person who decides how the accumulated funds are to be spent) and security (even his welfare interests might be threatened by the act that invades his financial interest, especially if the invasive act employs force or coercion, or seems likely to be frequently repeated).[81]

Coupled with the threat to liberty and security interests, even minor setbacks to the financial interests of others threaten “the general security of property, and the orderliness and predictability of financial affairs in which everyone has an interest, however small.”[82] Those security interests that cushion our welfare interests can be protected.[83] For instance, common assaults are criminalized to protect our elementary sense of security.[84] In a similar vein, our accumulative interests are those “nonessential interests [we] have in the various good things of life.”[85] The theft of a billionaire’s yacht or Caravaggio would not necessarily deprive a billionaire of their livelihood or margin of security above the minimum they require, but it would invade their accumulative resources.[86] If left unchecked, theft would also destabilize the entire property system in which we all have an interest.

Feinberg distinguishes mere wants from cognizable interests. It would be implausible to classify strong wants as interests. For example, Lucy, a devoted fan of the Yankees, may have a fervent desire to see the Yankees win, but that alone would hardly ground a case for claiming an interest in a Yankees victory.[87] Feinberg argues that “[s]ome of our most intense desires then are not of the appropriate kind to ground ulterior interests since (like a sudden craving for an ice cream cone) they are unlinked to our longer-range purposes, or they are insufficiently stable and durable to represent any investment of a stake.”[88] The harm principle is a measure that helps protect personal autonomy.[89] A person is harmed when their opportunities for enjoying or pursuing the “good life” are thwarted or diminished.[90] Harm occurs when our personal or proprietary resources are impaired, since our resources are needed to enable us to realize our other opportunities.[91]

Feinberg’s formulation of the harm principle has its problems, but it generally provides a fairly convincing account of the wrongness of certain harms both in the primitive and conventional sense.[92] If we return to the example of X lightly scratching the paintwork on Ys Rolls-Royce, we can see that the bad consequence is conventionally constructed as it does not necessarily damage Ys livelihood—nor does it automatically impact Y’s accumulative interests. It only impacts Y’s accumulative interests because Y has been socialized to perceive the scratch on the car’s original paintwork as an act of vandalism. It is the way in which Y has been socialized that causes Y to be offended by the aesthetics of the altered paintwork. Y is wronged because X interfered with Y’s autonomy (freedom of choice) by scratching the car without consent; and the wrong stands even though Y’s resentment hinges on the fact that Y has been socialized to dislike the car’s altered appearance. Y feels compelled to use Y’s accumulated resources to have the car restored to its original condition and thus gain control over how the property will be used. In most modern contexts, a car is not a necessity. Coupled with this, Ys car is still fully functional. Y’s accumulated primitive resources have not necessarily been diminished in the sense of overall essential livelihood such as basic food, shelter, physical security, and so forth[93] because Y can still gain full use from the car regardless of its altered paintwork. Feinberg draws on conventional property rights and argues that one should have the right to protect accumulated resources, even if those resources are unearned—as might be the case with celebrities who are often paid way beyond what a person could possibly earn for the labour and skill of a single human—because not having this right would be harmful in a normative sense. I think the protection of unearned wealth and excessive wealth surely has to rest on conventional notions of property rights.

Let us return to our example. Ys socialization will cause Y to feel compelled to have the paintwork on the Rolls-Royce restored to its original form. In fixing the paintwork, Y is asserting a right to decide how to use his or her property. Y will be harmed to the extent that Y will have to draw on accumulated resources in order to restore the car. Conventionally, it is arguable that it would be impossible for many people to live together co-operatively and seek the benefit of co-operative living without also accepting reasonable compromises. A person is expected to accept a compromise to their freedom when the exercise of that freedom has avoidable and unjustifiable, or, inexcusable bad consequences for others, even if those bad consequences are conventionally contingent. People queue when they are waiting to be served at the grocers or bank as this compromise allows each queue member to benefit from a fair distribution of the burdens and benefits that arise from co-operative living. The criminal law is used in more serious cases to coerce those who are unwilling to accept reasonable compromises.[94] If everyone were allowed to scratch the cars of others, this would unnecessarily threaten the general co-operative system of living (community and society) in which everyone has an interest. By scratching the car, X does not act in a way that is acceptable to rational agents who are trying to work together in a co-operative system.[95]

Criminalization, in this context, is designed to facilitate a fair distribution of the benefits and burdens that arise from co-operative communal living. The harm-doer is not asked to accept an unreasonable compromise in freedom as the freedom to wantonly invade the property rights of others is not a fundamental freedom and is thus an avoidable violation of the freedom of others. The underlying social norms play a significant role in explaining why a person might feel compelled to maintain the physical appearance of a car and thereby suffer a setback in accumulated resources in order to restore their property.

IV. The Conventional Badness of Offence-Doing

The conventional contingency of badness and, ultimately, wrongness is even more evident in the case of offence to others as its badness varies depending on the conventions adopted by the given community.[96] Feinberg asserts that “[i]t is always a good reason in support of a proposed criminal prohibition that it would probably be an effective way of preventing serious offense (as opposed to injury or harm) to persons other than the actor, and that it is probably a necessary means to that end.”[97] Feinberg argues that a separate offence principle is needed because ephemeral annoyances, disappointments, disgusts, embarrassments, and detested conditions such as fear, anxiety, and trivial aches and pains, do not necessarily result in harm. Some offensive encroachments might set back our interests and thus come within the purview of the harm principle, but most forms of offence do not result in harm. Even gross offences such as public displays of earrings made from human foetuses,[98] eating vomit in front of others in the confines of a public bus, copulating in public spaces, and so forth, do not amount to harm. Since

“harm” even in the broad untechnical sense rules out mere transitory disappointments, minor physical and mental “hurts,” and a miscellany of disliked states of mind, including various forms of offendedness, anxiety, and boredom as harms, since harm in the broad sense is any setback of an interest, and there is (typically) no interest in the avoidance of such states.[99]

Feinberg uses the concept of culpable offence to provide a critical moral reason for criminalizing those harmless but offensive acts that are “offenses proper (e.g., revulsion and disgust), hurts (e.g., ‘harmless’ throbs and pangs), and ‘others’ (e.g., shame and embarrassment).”[100]

We need to examine two questions: (1) what makes offending others wrong, and (2) what are the criminalizable bad consequences of this type of wrongdoing? Offence is assessed almost entirely in accordance with community mores, which differ from community to community, from generation to generation. For example, public nudity is not universally offensive. There is plenty of evidence that, in the history of the world, many cultures existed where public nudity was not an issue, and where privacy has not been conceptualized in the same way as it is in modern times. Even during more recent times in some remote tribes in tropical regions, it is common for people to go unclothed.[101] Public nudity is also common on many European beaches. Similarly, a local swearing at another local in an inner-city neighbourhood would not have the same social meaning as a philosophy student swearing at a professor on the High Table at a Cambridge University formal dinner. The sight of two men or women kissing in public might cause profound affront in some parts of Russia, the Middle East, or even in some parts of the United States,[102] but might go unnoticed in London, New York, or Stockholm. Offence is predominantly a subjective sensation, some forms of communication will offend the old but not the young, some forms of communication will offend women but not men, and some displays will offend Christians but not atheists, and vice versa.

What makes disgust objectively wrong? Underlying the wrongdoing involved in some forms of offence-doing is the idea of disrespect. Respect in its non-Kantian formulation would provide neutral reasons for tolerance and compromise. As Rescher states, “Respect people’s sensibilities about the appropriate and acceptable appearance of fellow humans by conforming to established rules of proper modesty.”[103] Simester and von Hirsch, drawing on the idea of respect and consideration for others, argue that the wrongness of—and according to them, the ability to criminalize—offensive behaviour can be explained in terms of disrespect and inconsiderate actions.[104] They argue that everyone, as self-determined morally responsible human beings, has a right to be treated with a minimum degree of respect and consideration. According to their analysis, certain offensive acts are impermissible because they treat autonomous choosing agents with a gross lack of respect and consideration. “The wrongdoing requirement calls upon the proponent of criminalisation to put forward reasons why the conduct is wrong—namely, under our proposed account, why the conduct treats others with a gross lack of consideration or respect.”[105] However, von Hirsch and Simester provide a conventional case for criminalizing certain offensive acts. They refer to particular bad consequences such as insults, inverse-privacy and loss of anonymity, rather than offence per se, to explain disrespect and inconsideration. I think a better approach would have been for them to test the boundaries of privacy violations as a conventional bad act or consequence, and to argue that the union of this bad act with culpability equals wrongness.

Von Hirsch and Simester postulate that the ability to criminalize exhibitionism might be drawn from

Nagel’s conception of “reticence”, regarding obligations of mutual restraint concerning person’s private (and especially their intimate) sphere. Notions of reticence include an entitlement to privacy—to exclude others from one’s personal domain. But the obverse should also obtain: we are entitled not to be involuntarily included in the personal domain of others—particularly, to be spared certain intimate revelations. It is the wrongfulness of that involuntary inclusion that, arguably, makes exhibitionism a matter of treating others without consideration.[106]

Does the bad consequence of having your privacy violated provide a principled justification for invoking the criminal law? There are deep conventional understandings concerning privacy in Western society and there is no doubt that gross privacy violations can have bad consequences for those affected. If a man films up a lady’s skirt and thereafter posts the images on the Internet, it is not difficult to envisage the conventionally bad result that this will have for the victim. Any psychological distress would depend on the way the victim has been socialized to feel shame, as she would likely not complain if someone took a photo of her face. However, this type of conventionally contingent bad consequence is real and should be criminalized because it violates the victim’s freedom and the defendant has no great liberty in the interest of filming up the skirts of others.

Let us consider the idea of privacy as a conventional justification for invoking the criminal law. Privacy, as defined by Gavison, involves “three independent components: in perfect privacy no one has any information about X, no one pays attention to X, and no one has physical access to X.”[107] These elements of secrecy, anonymity, and solitude are interrelated and all form a part of the complex fabric of the concept of privacy. A person can suffer a loss of proximity and anonymity when an uninvited stranger sits at that person's table in a restaurant, or next to them on the train even though the carriage is full of empty seats. This reference to physical access and physical proximity is in the sense of a person gaining the sort of access that would allow them to get close enough to touch or observe the captive viewer through the normal use of their senses.

Gavison provides a number of examples to demonstrate how certain privacy losses can be understood as physical access. For example, if “a stranger ... gains entrance to a woman’s house on false pretenses in order to watch her giving birth,”[108] it is the proximity violation that causes the loss of privacy. Similarly, if “a stranger ... chooses to sit on ‘our’ bench, even though the park is full of empty benches,”[109] it is the proximity violation and physical access that causes the loss of privacy in this context. In both of these cases, “the essence of the complaint is not that more information about us has been acquired or that more attention has been drawn to us, but that our spatial aloneness has been diminished.”[110] The context and underlying social norms are important. A person would not violate another’s privacy by standing right next to them in a crowded train. Two people may be forced to sit next to each other on a crowded train, but they do not violate the other’s territory as they have a conventional understanding about the right to sit near each other in this context. It is the norm for people to use the shared space in a public train. If the train is totally crowded, then people are expected to sit and stand closely together. The passengers expect this from experience. A person does not subject a fellow passenger to unwanted attention simply by sitting next to them.

I now want to turn my attention to the second and third types of privacy losses referred to by Gavison—that is, “no one pays attention to X, and no one has physical access to X.”[111] What are the bad consequences of exhibitionism? Take the example provided by Feinberg of the couple who copulate in a public bus.[112] People have a right to be able to copulate, but should members of the public be forced to see the intimate details of the couple’s copulation in a public place. Unwanted information such as nude displays, are obtrusions into domains.[113] Goffman notes that wrongful encroachments can come about either through an intrusion or an obtrusion.[114]X might intrude into Y’s physical space by taking Y’s private information, or entering Y’s private home, or even by getting too close to Y in a public park. Meanwhile, a wrongful obtrusion comes about “when an individual makes what are taken as overextensive claims to personal space, incidentally encroaching on the personal space of those adjacent to him or her on areas felt to be public in the sense of being non-claimable.”[115] For instance, if X were to plug a speaker into an iPhone and play music at full volume while riding in a public bus, this would disturb the comfort of those who want silence or a reasonably minimal level of noise.

In the chapter entitled “The Territories of the Self”, Goffman defines a territory as a “field of things”, or a “preserve”, that individuals have claims over.[116] In the situational sense, the individual would have an entitlement to control, use, or possess the demarcated territory. In the egocentric sense there are “preserves which move around with the claimant, he being in the center.”[117] Territories are not determined by objective factors, but rather the determiners are contextual. Their contours have a socially determined variability and are defined according to “[s]uch factors as local population density, purpose of the approacher, fixed seating equipment, character of the social occasion, and so forth.”[118] Goffman’s territories of self are defined by contextual and conventional factors rather than by objective criteria. Accordingly, he defines personal space as “[t]he space surrounding an individual, anywhere within which an entering other causes the individual to feel encroached upon, leading him to show displeasure and sometimes to withdraw.”[119] The contours of personal space are generally determined according to social norms,[120] so whether there is an objective violation of privacy depends on context and convention. For example, if a person sits next to the only other passenger in the train car, that passenger might find this invasive. If the person who sits next to the only other passenger is of the opposite sex, this might add dimension to the passenger's concern and discomfiture.[121] This sort of harassment has the potential to violate the train passenger’s right to be let alone and remain anonymous. Likewise, if a man goes to an almost empty beach and sits within a foot of a young woman, his propinquity would violate her right to be let alone. He is in her private domain—her territory. This would be an unjustifiable invasion of her personal space, over which she has a claim.

Egocentrically, a person’s personal space moves with them, “[the person] being in the center.”[122] An individual is entitled to exclude others from their territory. If the beach was absolutely packed, then keeping a distance of a foot might go unnoticed as he is merely asking her to share the public beach, which is a permissible demand. Beach-goers consent to the crowding by making the decision to use the crowded beach and they share the common end of using the beach for recreational purposes. Goffman notes:

[O]n the issue of will and self-determination turns the whole possibility of using territories of the self in a dual way, with comings-into-touch avoided as a means of maintaining respect and engaged in as a means of establishing regard. And on this duality rests the possibility of according meaning to territorial events and the practicality of so doing. It is no wonder that felt self-determination is crucial to one’s sense of what it means to be a full-fledged person.[123]

The conventional badness of an inverse-privacy violation can be understood as an autonomy violation. The privacy (and autonomy) violation is objectively bad to the extent that Western society values privacy and space. It is arguable that we all have an interest in maintaining a minimum degree of privacy, given the conventional and socialized makeup of moral agents in complex plural societies.

Unreasonable losses of autonomy occur when the wrongdoer’s private information is forced upon non-consenting spectators in the public domain. The bad consequence of exhibitionism is simply that the witness is denied the opportunity to choose whether to receive this very intimate information. The unwanted information could be obtrusive without necessarily having an obscene or indecent content. For example, if a person plays an iPod in the confines of a public bus at its highest volume, they are intruding on the personal domains of the other passengers.[124] The loud decibels resonating from the iPod in the public bus would restrict the choices of the other passengers by preventing them from choosing between loud music and silence, or between the loud music and conversation with other passengers. Feinberg argues that, “In being made to experience and be occupied in certain ways by outsiders, and having had no choice in the matter whatever, the captive passengers suffer a violation of their autonomy.”[125]

However, because the unwanted consequence is of a trivial nature, any criminal regulation should be enforced with fines rather than jail terms. The justification for criminalizing this type of wrongdoing has nothing to do with critical morality. Even the conventional case for criminalization is exceptionally weak. There is no deep agreement about the need to limit offensive behaviour in public places or about what is offensive. Many people are not concerned about exhibitionism in open public places such as nude beaches and parks. If a person goes to a nude beach, that person consents to what they might see. Von Hirsch and Simester argue that if the offence is readily avoidable, then it should not be criminalized.[126] Certain types of offensive conduct violate the autonomy and privacy rights of non-consenting audiences in public places, so they are subject to a ready avoidability requirement. That is, such activities will be criminalized when others are not able to readily avoid them without unwarranted restriction of their own liberty.[127]

Similarly, a loose conventional argument for criminalizing traditional privacy violations might go as follows: due to social conditioning, citizens living in Western societies have deep feelings about privacy. Revealing another person’s very intimate information could cause them grave distress because of their social beliefs and the shaming norms in our society. People are made to feel ashamed of certain things and therefore require privacy. Privacy violations are clearly only bad in a conventionally contingent sense, but may cause genuine psychological distress. Modern technology has made it particularly easy to access and distribute private information via the Internet. In some cases, computer repairmen have discovered very private information, such as obscene images, on computers belonging to their celebrity customers and have uploaded the information onto the Internet without the customer’s consent.[128] In these cases the privacy violation is grave enough to result in conventionally contingent harm. In perfect privacy, no one has any information about the non-consenting party. There is, however, no such thing as perfect privacy. We all give up some privacy by entering the public domain to do our daily business, and entering the public domain means public surveillance. We have deep conventional understandings about privacy, and a person is entitled to keep certain intimate information private regardless of whether or not they are a public figure.

When a person uploads obscene images of non-consenting adults onto the Internet, this causes a grave loss of privacy. This type of direct privacy violation could come within the purview of the criminal law when the material is of an exceptionally private and intimate nature. The material would have to be exceptionally intimate in those cases involving public figures because they gain benefits from public prominence and must expect more public scrutiny than the average citizen. If a person puts themself in the public spotlight, then they have to expect a much greater degree of media scrutiny than non-celebrities receive. Culpably distributing photos and movies of celebrities, or of ex-partners copulating without their consent, is prima facie criminalizable because it has bad consequences for the non-consenting party. Trivial or moderate privacy violations should not be criminalized, as the civil law provides adequate remedies.[129]

There are limitations on the types of information that can be disclosed without consent. When a person distributes private images of a celebrity (or of anyone) that is of an obscene nature, the public has no interest in seeing it, and the non-consenting party is entitled to have their privacy protected by use of the criminal law. It is one thing to report that a public figure is having a love affair, but it would be something entirely different to publish private images of that person having actual intercourse or of that person in the nude. It would not be reasonable to expect the affected party to personally undertake responsibility for proceedings to put a stop to it. The criminal law protection of privacy cannot be extended to other forms of intimate information (i.e., written or oral mention of the sensitive information) involving celebrities, because this would be too great a restriction on freedom of expression and too great an extension of the criminal law. Mere verbal or written mention, or non-obscene images, of a celebrity love affair would not be enough to justify invoking the criminal law.

Similarly, tacit consent will be sufficient to waive the right to privacy. If a celebrity (or anyone else for that matter) goes to a public nude beach, then the celebrity is making their intimate information public. If the paparazzi takes photos or movies of the display and distributes it to the tabloids, then the affected party cannot claim that their privacy has been violated because the affected party has made the information public by presenting it in the public domain. There is, however, a distinction between public and quasi-public places. If a journalist or photographer gains access to a locker room and takes a photo of a famous actor or footballer in the nude, the photographer could hardly claim that the victim tacitly consented to this intimate information being made public.[130] Where a celebrity intentionally or indifferently exposes their posterior in public, the celebrity cannot complain of a loss of privacy.[131]

V. Principled Criminalization and Conventionally Contingent Wrongs

Given that the objective badness of privacy loss is conventional, is it possible to build a principled case for criminalizing such wrongs? Some remote groups have been socialized not to value privacy or to perceive public nudity as intimate and unwanted information.[132] Unlike the bad consequence of harm to others—basically an uncontested bad consequence in the primitive sense of harm—there is much less agreement about the badness of many forms of privacy violations. Measuring the justice of criminalizing conduct exclusively in terms of its impact on human values and experience does not mean that we cannot distinguish unprincipled criminalization from principled criminalization. It is by drawing on those values and experiences that we are able to identify conduct that is worthy of criminalization. Feinberg’s harm principle can be grounded on a conventionalist account of harm and culpability. What we are able to do is draw on sociological and scientific evidence to make our harm claims objective within the conventional paradigm.

There is no need to claim that our conceptualizations of harm and offence are correct according to some universal standard. Our harm and offence claims can be tested against our core conventional and scientific understandings of harm. Harm is universal in the sense that all cultures recognize harm as a bad thing. Where societies differ is not on the notion that harm is bad, but on what counts as harm. At the base point, there are primitive harms that empirically and biologically affect all humans in the same way. For instance, torture will result in physical pain regardless of where the victim might be culturally situated. When a rogue state uses torture to obtain information from captured soldiers, it does so because torture is bad and harmful—if it were not the captors would use some other method to obtain the information. Rewards such as bribes work the opposite way. Once we move away from the most primitive harms, it becomes more difficult to identify the types of harms that would ground principled criminalization.[133] We have conventional understandings about the psychological distress caused by grossly offensives acts, but it is more difficult to supply principled justifications for criminalizing them. But if our conventional justifications are able to withstand the detached scrutiny of communally situated agents, they will be relatively principled.

In Western societies, privacy is considered to be a cardinal value. Privacy violations in the traditional sense, present a stronger case for principled criminalization as the victim might be left rather traumatized by having their private acts uploaded onto the Internet. Due to the social makeup of citizens living in competitive and sophisticated modern societies, privacy has evolved as a cardinal want. When X uses a hidden camera to film up Y’s skirt,[134]X violates Y’s autonomy by deciding how X’s private information will be used. Stanley I. Benn rightly argues that such a violation is wrong because it treats its victim with a lack of respect as a person.[135] Benn argues that covert surveillance is morally wrong because it “deliberately deceives a person about his world: It thwarts, on the basis of reasons that are not his own, the agent’s attempts to make rational choices.”[136] This type of violation is wrong even though the information (movie, photos, and so forth) might never be made public, not merely because the clandestine spying would hurt the victim’s feelings, but because the wrongdoer uses the unsuspecting victims as a mere means to serve the ends of the wrongdoer. Keeping the spying secret so that the victims do not find out might inadvertently spare the victims’ feelings, but it would also add dimension to the wrongness of the spying because it falsifies the victim’s self-perception. The victims might, acting on the false belief that they are in control of their private world, act even more “intriguingly for [their] manipulator’s ends.”[137] Benn goes on to assert that, “One cannot respect someone as engaged on an enterprise worthy of consideration if one knowingly and deliberately alters his conditions of action while concealing the fact from him.”[138] Benn’s formula is Kantian, but his account of wrongful privacy violations encapsulates our conventional understanding of the wrongness of privacy violations fairly well.

Those who learn about the covert spying would feel resentment and would be offended within Feinberg’s wide definition of offence. The short-lived anger and psychological distress would not be enough to set back their interests, but the loss of privacy and anonymity would cause profound distress and resentment. This type of wrongdoing is criminalizable not only because it causes major distress, but because it also results in a culpable violation of the victim’s autonomy and privacy rights. The privacy loss is an independent bad consequence, which can be used to give the legislature guidance and justification for invoking the criminal law. It is a consequence that is intersubjectively accepted as bad by communally situated agents. What is also important is that the violator has no interest in revealing or accessing this type of private information from others. The distress caused by this type of violation is conventionally contingent since, in a culture where nudity or a lack of privacy is the norm for example, people might not care less if someone were to film them in a state of undress.

Nonetheless, the case for criminalization is exceptionally weak when the defendant is not revealing or accessing another’s private information, but is instead forcing their own private information on others. X might claim to have suffered a loss of autonomy when forced to receive offensive information in a public context because X is denied the opportunity to avoid receiving the unwanted information. The offended party could no doubt claim they were disgusted and shocked by the display of a copulating couple, but so might those who have been socialized to find same-sex couples disgusting. They might also claim that being confined in a public bus where same-sex couples are merely kissing is a bad consequence for them because it is something they do not want to see. What if a woman is topless on the bus, is this different from a man being topless on the bus? Such a law would violate our deeply held conventions about equality. Thus, if a law were to ban all kissing on public buses regardless of sexual orientation, it may be permissible if it were really needed to prevent a bad consequence. Similarly, a law would be discriminatory if it prevented only women from going topless in public.

What makes forcing members of the public to deal with offensive information wrong? The offence is subjectively taken in all cases because some passengers on a bus might be voyeurs and might not mind seeing copulating couples. Meanwhile, same-sex couples merely kissing would not offend many others. If it is just the union of culpability and the consequence of forcing the offended party to receive information that they do not subjectively want to receive that justifies criminalization,[139] then there is no way of dealing with the problem of unprincipled criminalization beyond referring to agreed upon core harms. I am of the view that, once we move away from culpable harm criteria, the case for criminalization becomes substantially weaker. Having said that, the gross privacy violations involved in uploading private information onto the Internet could come within the purview of the harm concept.[140] The latter is distinguishable because people have a stronger interest in protecting their own information. The interest in not receiving other people’s private information in public contexts is not great, as free speech is a much more cardinal liberty.

One justification for criminalizing the copulating couple on the public bus would be that their actions are unhygienic—other people want to use the public bus seats without sitting where the couple once did their act. Likewise, a person would not want to sit on a restaurant seat if a nudist had just been sitting on it. The harm principle might be invoked in such cases. Similarly, uncovered nudists would not be welcome around the buffet in a restaurant. This would be a soft harm. However, if these people were on an open beach or in an open park, the case for criminalization seems hard to sustain. Women wearing a burka in public places might cause offence to different people for different reasons,[141] but we would not want to use the criminal law to tell people how they should dress in public. Therefore, there is no strong case for criminalizing exhibitionism in open public places where food is not being served or where public seating is not involved. This probably explains why exhibitionism is not criminalized on beaches and in parks in many European countries.

To the extent that people claim that they have a right not to see same-sex couples kissing, women wearing burkas, nudists on a beach, and so forth, the criminal law has no role to play in this. Feinberg cites the reasonable nature of such offensive displays[142] as a reason for tolerance, but the better justification for tolerating such conduct is that it does not wrong others in a conventional sense because there is deep agreement about the need to tolerate diversity in modern societies. Likewise, the deep offence caused to some by knowing that books such as Salman Rushdie’s The Satanic Verses[143] or Philip Roth’s Portnoy’s Complaint[144] exist, does not provide a justification for outlawing such literature. To the extent that people are offended by simply knowing that such activities are taking place behind closed doors, or that such books exist, Murphy rightly notes:

We must remember that ex hypothesi the acts in question are performed in private by consenting adults. Thus the only thing to which the complainant could object is the bare knowledge that something of which he disapproves is going on in private. The question, then, is this: Is freedom from a knowledge that some disapproved activity is taking place a right that ought to be recognized? Hart argues convincingly that it is not.[145]

Unless a person is forced to watch or read the offensive film or book, they could hardly claim to have been wronged. Would denying the Cambodian genocide or the Nanking Massacre at Speaker’s Corner in London’s Hyde Park violate a passing survivor’s right not to receive such information? This type of speech clearly treats the passerby with a gross lack of consideration. But this type of political dialogue belongs in the public arena, even though it is factually wrong and offensive, because it is enlightening to the extent it allows the public to understand that some very disturbing views exist, and leads to informed debate, which allows the record to be set straight. People have an interest in knowing that these types of awful views exist in the real world, and they have a responsibility to publicly denounce such views.[146] This type of information is public information—not private information—and therefore does not violate the rights of those who are forced to receive it. Such receivers consent to receiving this information by choosing to visit public places. People give up a certain amount of privacy and autonomy as soon as they walk out of their front door, and in doing this, they cannot expect to be sheltered from the real world. Different political views are not the type of sensitive and private information that a person could claim a right to be sheltered from. Per contra, copulation in a public bus does not serve a similar purpose, and the intimate display is of no benefit to those who do not want to endure this type of up close and unavoidable encounter. But if it did occur the public would have an interest in hearing or reading about it in the news, as it informs them about what is going on in the world.

The freedom of expression right protects this interest by creating an environment that facilitates the free flow of information. Waldron points out that panhandlers cause offence to some passing by as they may be distressed by the message conveyed by homeless people, but that distress is beneficial for both the offended party and the homeless person.[147] Firstly, the passerby might say, “This is awful. I am glad I have found out about this.”[148] Additionally, the encounter might even motivate the passerby to do something about it. Such an encounter is a good consequence, not a bad consequence to the extent it is not aggressive. Aggressively targeting passersby is a form of harassment and would most likely come within the purview of the harm principle because it forces society to acknowledge and respond to the harsh realities of indigence.[149] Unwanted speech that captures the public’s attention is usually seen as a detriment, but it should not be seen as a detriment from the audience’s point of view. As John Stuart Mill is cited as rightly emphasizing, “[T]here is significant benefit in being exposed to ideas and attitudes different from one’s own, though this exposure may be unwelcome.”[150] The panhandler, the conservative, the liberal, and the prejudiced individual, all have an interest in articulating their views and in hearing the views of their adversaries. The consequence of being forced to deal with public political discussion is somewhat different than having a copulating couple’s live presentation of their private affairs forced upon you in the confines of a public bus.

Feinberg correctly asserts that offence would hardly ever outweigh the value of free speech.[151] It appears that only a very narrow range of very intimate displays (copulating in a bus, or physical intrusions such as accessing people for prolonged periods in certain contexts in public places, for example) would be sufficiently bad to justify a criminal law response. Such activities interfere with the auditor’s right to be let alone and their right not to be proximately included in the very private affairs of others in confined public contexts. However, only a weak case can be made for criminalizing exhibitionism more generally since the bad consequence is conventionally variable and could barely be said to have significant consequences for others when carried out in open spaces as opposed to confined public places.


There are established human values and conventions. There are also recognized standards of rational argument. John McDowell takes the view that moral values are both anthropocentric and real. He also argues that objectivity claims are to be made from the internal perspective of our actual practices.[152] We might draw on our best theories of thought, language, and so forth, but this does not really tell us whether our claims of wrongness are truly objective. In this paper I have argued that principled criminalization does not have to rely on critical objectivity in the sense of producing transcultural and truly correct standards. I have argued that Feinberg’s harm principle can be supported with conventional accounts of harm. The best that we can do is scrutinize our conventional conceptualizations of harm and badness, but that scrutiny is constrained by the limits of epistemological inquiry and our capacity for rationality at any given point in time. The conflicts that arise from communal living inevitably lead to some kind of political philosophy. Many acts are criminalizable because they violate conventions that are shareable by communally situated agents.

I have noted that there is no deep intersubjective agreement about the badness of exhibitionism and other similar examples. Nor is it clear that we have a shareable end in outlawing it. The offended group itself will only follow norms against it while continuing to collectively “maintain certain attitudes and beliefs concerning them.”[153] We might criminalize certain soft harms, such as privacy violations, to prevent those who have been socialized to value privacy from suffering humiliation and psychological distress. But if that distress arises from other forms of offence, such as exhibitionism or hate speech, the cardinal value of freedom of expression would most likely override the victim’s liberty in avoiding this type of offence. As noted above, offence can stimulate debate and tolerance. The difference with exhibitionism is that the offending party has an overriding liberty, unless the exhibitionism occurs in a context where it raises hygiene issues or targets a confined observer, as might be the case on a public bus. The bad consequence of loss of privacy is conventionally contingent—it is contingent on the way the victim has been socialized.

The gravity of sexual objectivity also hinges on socialization and convention. Take a society where people might be socialized from a young age to believe that it is a great honour to be used as a sex object by a senior member of society.[154] Obviously, in such a society the psychological harm would differ from that faced by victims of sexual abuse in modern Western society. What is undeniable is that socialization means that sexual abuse not only results in physical harm in modern societies, but also genuine psychological harm and trauma. There is a consensus about the need to criminalize sexual abuse because of its psychological and physical badness. Due to the way people are socialized, some forms of sexual use seem to be tolerated in modern society. For instance, it is arguable that to some extent people are socialized to tolerate sexual use from celebrities and other powerful figures. It is doubtful that Tiger Woods, a celebrity golfer, would have managed to convince so many women to participate in his infidelity if he had been a normal labourer rather than a skilled labourer.[155] As J. L. Mackie notes:

Only some kinds of harm are socially, cooperatively, resented, and cooperation in gratitude is even more restricted. Again we must seek and can find sociological reasons for these differences: only with particular kinds of harm are the conditions favourable for the growth of a convention of cooperative hostility to them, so only some kinds of harm are seen as wrong and as calling for general resentment and punishment. ... Though retributive principles cannot be defended, with any plausibility, as allegedly objective moral truths, retributive attitudes can be readily understood and explained as sentiments that have grown up and are sustained partly through biological processes, and partly through analogous sociological ones.[156]

Even our best sociological and biological conceptualizations of harm do not cover mere umbrage. We might invoke Lord Devlin’s argument and hold that if the majority, “do not like it then criminalize it,” but that would allow anything to be criminalized. The argument for the preservation of social harmony is weaker in the case of personal space violations than it is in the case of those violations that are likely to have deeper psychological consequences, such as sexual abuse or filming up a lady’s skirt.[157] When the conduct involves soft harms, the criminal law should be used only as a last resort.

I have aimed to identify some of the preliminary issues that have to be sorted out before more work can be done. Some might argue that, apart from primitive harms common to all societies, my criterion of wrong is whatever culturally situated agents consider as intersubjectively wrong. The question then is how my theory of badness differs from positive morality. It does not—except that it requires us to scrutinize the justifications we put forward for criminalizing soft harms and offensive acts. I would hope that modern thinkers would subject our conventional practices to greater scrutiny than Lord Devlin did. Finally, I note that the problem for those who want transcultural accounts of harm and offence, is that procedural realism is not convincing as a method for grounding such standards.[158] Procedural realism has greater potential as a mechanism for scrutinizing conventional standards, but it would have to accept that the deliberators are merely human agents who are communally situated.

It is possible to have a principled criminal law, but this would mean accepting deeply held conventional accounts of harm, or soft harm, and offence. It would also mean accepting that something that is considered as a soft harm now, might not be considered as one in the future or in some other cultural context. There is a clear case for criminalizing harms of a more primitive kind. There are strong conventions telling us not to rape, murder, steal, and so forth. There are also clear conventional understandings about the humiliation and psychological distress that might flow form privacy violations. It is essential for us to scrutinize soft harms and offensive acts much more than is necessary for primitive harms, and in undertaking this scrutiny we should be able to reduce unprincipled criminalization. This cannot provide a perfect solution, but it does provide more guidance than critical morality.